Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
6 results
Search Results
Item Multistep ahead groundwater level time-series forecasting using gaussian process regression and ANFIS(Springer Verlag service@springer.de, 2016) Naganna, N.S.; Deka, P.C.Groundwater level is regarded as an environmental indicator to quantify groundwater resources and their exploitation. In general, groundwater systems are characterized by complex and nonlinear features. Gaussian Process Regression (GPR) approach is employed in the present study to investigate its applicability in probabilistic forecasting of monthly groundwater level fluctuations at two shallow unconfined aquifers located in the Kumaradhara river basin near Sullia Taluk, India. A series of monthly groundwater level observations monitored during the period 2000–2013 is utilized for the simulation. Univariate time-series GPR and Adaptive Neuro Fuzzy Inference System (ANFIS) models are simulated and applied for multistep lead time forecasting of groundwater levels. Individual performance of the GPR and ANFIS models are comparatively evaluated using various statistical indices. In overall, simulation results reveal that GPR model provided reasonably accurate predictions than that of ANFIS during both training and testing phases. Thus, an effective GPR model is found to generate more precise probabilistic forecasts of groundwater levels. © Springer India 2016.Item Classification of case-II waters using hyperspectral (HICO) data over North Indian Ocean(SPIE spie@spie.org, 2016) Srinivasa Rao, N.; Ramarao, E.P.; Srinivas, K.; Deka, P.C.State of the art Ocean color algorithms are proven for retrieving the ocean constituents (chlorophyll-a, CDOM and Suspended Sediments) in case-I waters. However, these algorithms could not perform well at case-II waters because of the optical complexity. Hyperspectral data is found to be promising to classify the case-II waters. The aim of this study is to propose the spectral bands for future Ocean color sensors to classify the case-II waters. Study has been performed with Rrs's of HICO at estuaries of the river Indus and GBM of North Indian Ocean. Appropriate field samples are not available to validate and propose empirical models to retrieve concentrations. The sensor HICO is not currently operational to plan validation exercise. Aqua MODIS data at case-I and Case-II waters are used as complementary to in- situ. Analysis of Spectral reflectance curves suggests the band ratios of Rrs 484 nm and Rrs 581 nm, Rrs 490 nm and Rrs 426 nm to classify the Chlorophyll -a and CDOM respectively. Rrs 610 nm gives the best scope for suspended sediment retrieval. The work suggests the need for ocean color sensors with central wavelength's of 426, 484, 490, 581 and 610 nm to estimate the concentrations of Chl-a, Suspended Sediments and CDOM in case-II waters. © 2016 SPIE.Item Forecasting of Significant Wave Height Using Support Vector Regression(Institute of Electrical and Electronics Engineers Inc., 2016) Ajeesh, K.; Deka, P.C.The reliability of wave prediction is a crucial issue in coastal, harbor and ocean engineering. Support vector machine (SVM) is an appropriate and suitable method for significant wave height (Hs) prediction due to its best versatility, robustness, and effectiveness. In this present work, only significant wave height (Hs) of previous time steps were used as predictors during the period 01-01-2004 to 01-04-2004. The data used is processed significant wave height (Hs) of the station SW4(Latitude 12056′31″ and longitude 74043′58″) located near west coast of India.70% of the data used for calibration of model parameters and remaining 30% data used for validation using various input combinations. The performance of both the RBF and PUK models is assessed using different statistical indices. (E.g. CC (RBF - SVR) = 0.82, CC (PUK-SVR) = 0.93, MAE (RBF - SVR) = 0.04, MAE (PUK-SVR) =0.04 RMSE (RBF-SVR) =0.06, RMSE (PUK-SVR) =0.05. The results show that SVM can be successfully used for prediction of Hs. © 2015 IEEE.Item Identification of potential sources affecting fine particulate matter concentration in delhi, india(Springer Science and Business Media Deutschland GmbH, 2021) Harsha, K.; Shiva Nagendra, S.M.S.; Deka, P.C.The seasonal air transport pathways and the potential sources contributing to air pollution in Delhi for the period March 2015 to February 2016 have been identified with the help of PM2.5 data (particulate matter with diameter less than 2.5 μm), potential source contribution function (PSCF), cluster analysis and concentration weighted trajectory (CWT) method. The local sources are identified with the help of conditional probability function (CPF). The presence of re-circulating air masses has shown that the major contributors to air pollution in winter seasons are the local sources and the neighboring states of Haryana and Punjab. Northwesterly flows can be observed throughout the year and are highest in the winter season and comparatively lower in the monsoon season. PSCF values greater than 0.7 and CWT values greater than 110 μgm−3 are observed within the state in the winter season. Haryana and some parts of Uttar Pradesh also have higher PSCF values. The frequency of occurrence of long distance pathways is less in all the seasons in Delhi. The influence of the dust pathways from the Thar Desert areas can be seen in the monsoon season. Slower moving northwesterly and southwesterly flows are associated with high concentration values and indicate high pollution along the pathways. Higher CPF values occur in the northeastern direction. Therefore, the industrial sites, traffic congestion and emission from vehicles in the roads connecting Delhi and Uttar Pradesh have high influence in the rise in pollution levels. © Springer Nature Singapore Pte Ltd 2021.Item Forecasting of Meteorological Drought Using Machine Learning Algorithm(Springer Science and Business Media Deutschland GmbH, 2022) Kikon, A.; Deka, P.C.Drought forecasting is one of the crucial tools for the water management system, and understanding the different climatic variables affecting the occurrences of drought is a major scientific challenge. In this study, drought forecasting is done for the Peninsular region of India using different machine learning algorithms. A meteorological drought index known as Standardized Precipitation Index (SPI) which is dependent on the precipitation is taken into account for analysis. The SPI with a different timescale for 3-, 6-, 9-, 12-month were calculated from 1958–2017 for 60 years. SPI is a function of precipitation and the trend of rainfall followed may be found to be similar in some regions. Two different models, GA-ANFIS and GRNN were compared in this study. The results obtained from the statistical performance assessment of the models were compared with each other. For different timescale, there is a variation in its evaluation metrics. Comparing the performance assessment of the two different models, it is noticeable that the performance assessment of the statistics of the GA-ANFIS model outperformed GRNN model. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item Spatial mapping of travel information and assessment of road connectivity(International Society for Photogrammetry and Remote Sensing, 2023) Ashritha, K.; Deka, P.C.Roads play a crucial role in the urban spatial structure. A place's development and growth depend on the road network connectivity and accessibility being the socio-economic and transportation carrier. It involves the mobility of people and goods from one place to another. The choice of mode of travel depends on the living standards, connectivity, and vicinity to the work area. The study uses satellite data to analyze road network connectivity using the connectivity indices of Mangalore City Corporation, a port city in India. The connectivity indices alpha, beta, gamma, and eta showed the Area's good connectivity with proper roads and interconnectivity. Using Dijkstra's algorithm, the least cost path is identified on which the spatial mapping of the travel information is made. The travel information raster served the commuter in knowing the time, distance, and cost of modes from all possible origins to each city center. Specifically, it serves as the base map for bus routes, their cost, and travel time for significant city bus stations. The cost of travel, Duration, and distance information is mapped for two-wheeler and four-wheeler commuters. The study used the Modis Land Use Land Cover Data to identify inaccessible road network areas. © 2023 International Society for Photogrammetry and Remote Sensing. All rights reserved.